A Chloroplast DNA Phylogeny of Solanum Section Lasiocarpa
A Chloroplast DNA Phylogeny of Solanum Section Lasiocarpa
A Chloroplast DNA Phylogeny of Solanum Section Lasiocarpa
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
186 SYSTEMATIC BOTANY<br />
[Volume 29<br />
al. (1995) in their analyses <strong>of</strong> cp<strong>DNA</strong> restriction sites<br />
and combined cp<strong>DNA</strong>, morphological, and isozyme<br />
data. However, their analyses <strong>of</strong> morphological and<br />
isozyme data, alone and in combination, placed S. hirtum<br />
within this clade, whereas this species forms an<br />
isolated basal branch in the trn trees. Data from crossing<br />
studies and karyotype analyses are equivocal with<br />
respect to support for this large group (Heiser 1972,<br />
1989; Bernardello et al. 1994).<br />
Within this large clade, coded indel data provide<br />
some support for the association <strong>of</strong> S. hyporhodiumwith<br />
S. felinum. In analyses without the coded indel data, all<br />
accessions <strong>of</strong> these two species along with S. vestissimum<br />
form a basal grade in the large clade described<br />
above, with S. vestissimum S432 comprising the basal<br />
branch in the entire large clade. All three <strong>of</strong> these taxa<br />
are high-elevation cloud forest species native to Venezuela<br />
and northern Colombia. Whalen et al. (1981) considered<br />
the three species to be closely related on morphological<br />
grounds. <strong>Solanum</strong> felinum and S. vestissimum<br />
are extremely similar morphologically, with S. hyporhodium<br />
less so (Bruneau et al. 1995). <strong>Solanum</strong> hyporhodium<br />
and S. vestissimum clustered together in phenetic and<br />
cladistic analyses <strong>of</strong> isozyme data (Whalen and Caruso<br />
1983; Bruneau et al. 1995); S. felinum was not included<br />
in these studies. Crossing and karyotypic studies did<br />
not support a relationship among the three taxa (Heiser<br />
1972, 1989; Bernardello et al. 1994), although S. hyporhodium<br />
and S. felinum had similar chromosome characteristics<br />
(Bernardello et al. 1994). Although the three<br />
taxa are closely associated in most <strong>of</strong> the trn trees, they<br />
do not form a monophyletic group. In addition, the<br />
S432 accession <strong>of</strong> S. vestissimum from Colombia is divergent<br />
from the other four representatives <strong>of</strong> the<br />
group, all <strong>of</strong> which are from Venezuela. Heiser (2001)<br />
noted that accessions identied as S. vestissimum from<br />
Colombia and Venezuela would not cross with each<br />
other and differed in their crossing behavior with S.<br />
quitoense. Further taxonomic work on species limits in<br />
this complex and more intensive sampling with more<br />
variable genes is warranted to ascertain the position <strong>of</strong><br />
these high altitude Colombian and Venezuelan taxa.<br />
Two questions that have been intensively studied<br />
with respect to this group <strong>of</strong> species concern the wild<br />
relatives <strong>of</strong> S. quitoense and the origin and relationships<br />
<strong>of</strong> the two Old World taxa <strong>of</strong> section <strong>Lasiocarpa</strong>. <strong>Solanum</strong><br />
quitoense, the lulo or naranjilla, is a commonly<br />
cultivated fruit crop in Andean South America. Its<br />
range has recently spread to include Central America,<br />
where it is naturalized in Panama and Costa Rica. <strong>Solanum</strong><br />
quitoense has been considered by some to be<br />
known only from cultivation, although spiny and feral<br />
forms exist in northwestern South America. Heiser<br />
(1972) proposed on morphological grounds that S. quitoense<br />
is most closely related to S. candidum, but the<br />
two species have different habitat preferences and hy-<br />
bridize only with difculty. Although S. quitoense and<br />
S. candidum are not sister taxa in the trn trees, there is<br />
little character support and resolution in this area <strong>of</strong><br />
the tree and a close relationship between the two taxa<br />
cannot be ruled out. However, the trn data refute hypotheses<br />
<strong>of</strong> close associations between S. quitoense and<br />
S. hirtum, S. pectinatum, S. stramonifolium, and S. sessiliorum.<br />
Likewise, the relationships <strong>of</strong> the two Asian disjuncts,<br />
S. repandum and S. lasiocarpum, have been a matter<br />
<strong>of</strong> debate. Whalen et al. (1981) and Whalen and<br />
Caruso (1983) suggested that S. repandum and S. lasiocarpum<br />
were not sister taxa, but instead that S. repandum<br />
was allied to and perhaps conspecic with S. sessiliorum,<br />
whereas S. lasiocarpum was most closely related<br />
to S. candidum. Conversely, Heiser considered S.<br />
repandum and S. lasiocarpum to be closely related and<br />
perhaps conspecic (as S. ferox) and that S. candidum<br />
was sister to the Asian taxa (Heiser 1986, 1987, 1996).<br />
The trn data, as well as previous data from crossing<br />
and karytotype studies and analyses <strong>of</strong> cp<strong>DNA</strong> and<br />
morphological characters (Heiser 1986, 1987, 1996; Bernardello<br />
et al. 1994; Bruneau et al. 1995) supports the<br />
close relationship between S. repandum and S. lasiocarpum<br />
and thus Heiser’s hypothesis. Furthermore, S. candidum<br />
emerges as a member <strong>of</strong> the S. repandum/S. lasiocarpum<br />
clade, conforming to Heiser’s ideas <strong>of</strong> relationships.<br />
However, S. repandum and S. lasiocarpum did<br />
not form a monophyletic group in the trn analyses;<br />
rather, one accession <strong>of</strong> S. repandum formed a clade<br />
with the two S. pseudolulo accesssions. This result<br />
should not be over-interpreted, however, since there is<br />
little character support for the identication <strong>of</strong> lineages<br />
within the large clade that includes S. repandum, S. lasiocarpum,<br />
S. pseudolulo, S. candidum, S. quitoense, S. hyporhodium,<br />
S. vestissimum, and S. felinum.<br />
In general, the trn trees are quite similar to those<br />
obtained from analyses <strong>of</strong> cp<strong>DNA</strong> restriction site data<br />
(cf. Fig. 1 in Bruneau et al. 1995). This is not surprising,<br />
given that the chloroplast genome is a single linked<br />
non-recombining genetic entity (Doyle 1992). Further<br />
molecular studies are underway using more variable<br />
nuclear genes in order to achieve better resolution <strong>of</strong><br />
phylogenetic relationships among the species <strong>of</strong> section<br />
<strong>Lasiocarpa</strong>, to increase support for previously identied<br />
clades, and to compare phylogenies derived<br />
from maternally inherited chloroplast genes with those<br />
based on biparentally inherited nuclear markers.<br />
ACKNOWLEDGMENTS. I thank C. Heiser, A. Bruneau, J. Miller,<br />
P. Diggle, R. Olmstead, L. D. Gómez, and the Botanic Garden at<br />
the University <strong>of</strong> Nijmegen, The Netherlands, for providing seed<br />
and <strong>DNA</strong> samples; B. Hammel and R. Aguilar for eld assistance;<br />
R. Dirig for information regarding vouchers at BH; C. Heiser, A.<br />
Bruneau, E. E. Dickson, J. Miller, P. Diggle, S. Knapp, and D. Spooner<br />
for comments on the manuscript and help with sample and<br />
voucher information; M. Johnson, A. Freeman, S. King-Jones, A.<br />
Egan, A. Moore, and K. Klatt for laboratory assistance; and C.